Jet spectral breaks in black hole X-ray binaries

In X-ray binaries, compact jets are known to commonly radiate at radio to infrared frequencies, whereas at optical to ?-ray energies, the contribution of the jet is debated. The total luminosity, and hence power of the jet, is critically dependent on the position of the break in its spectrum, between optically thick (self-absorbed) and optically thin synchrotron emission. This break, or turnover, has been reported in just one black hole X-ray binary (BHXB) thus far, GX 339-4, and inferred via spectral fitting in two others, A0620-00 and Cyg X-1. Here, we collect a wealth of multi-wavelength data from the outbursts of BHXBs during hard X-ray states, in order to search for jet breaks as yet unidentified in their spectral energy distributions. In particular, we report the direct detection of the jet break in the spectrum of V404 Cyg during its 1989 outburst, at ?_b = (1.8 ± 0.3) × 10^14 Hz (1.7 ± 0.2?µm). We increase the number of BHXBs with measured jet breaks from three to eight. Jet breaks are found at frequencies spanning more than two orders of magnitude, from ?_b = (4.5 ± 0.8) × 10^12 Hz for XTE J1118+480 during its 2005 outburst, to ?_b > 4.7 × 10^14 Hz for V4641 Sgr in outburst. A positive correlation between jet break frequency and luminosity is expected theoretically; ?_b?L~ 0.5?_jet if other parameters are constant. With constraints on the jet break in a total of 12 BHXBs including two quiescent systems, we find a large range of jet break frequencies at similar luminosities and no obvious global relation (but such a relation cannot be ruled out for individual sources). We speculate that different magnetic field strengths and/or different radii of the acceleration zone in the inner regions of the jet are likely to be responsible for the observed scatter between sources. There is evidence that the high-energy cooling break in the jet spectrum shifts from UV energies at L_X ~ 10^-8 L_Edd (implying the jet may dominate the X-ray emission in quiescence) to X-ray energies at ~10^-3 L_Edd. Finally, we find that the jet break luminosity scales as L?, jet?L^0.56 ± 0.05_X (very similar to the radio–X-ray correlation), and radio-faint BHXBs have fainter jet breaks. In quiescence the jet break luminosity exceeds the X-ray luminosity